“We trained models on internal data and then predicted on external data… total number of tests … like 13 million across 500 protein targets.”
IQ
Ian Quigley
Latent Space: The AI Engineer Podcast ·
“We discovered this by accident... we can use it as one filter... if we get a bunch of Luma-Cafor heads, that protein probably didn't fold right. We probably don't want to use that as training data.”
Latent Space: The AI Engineer Podcast · Lab-in-the-Loop at Compute Speed - Ian Quigley, Leash Bio
Play from 1:45:35
1:45:35
Audio streams from the publisher’s own feed. Verbatim does not host or copy it.
What was described
used the presence of Luma-Cafor-like hits as a proxy that a protein sample misfolded, and excluded those data from model training
Said nearby on the same subject
“You need more pre-training data… give the model a bunch of data and think… transform this data… pre-trained on it would make the model smarter”
♫
Said on "The Cognitive Revolution"
“We assembled the largest possible data set… 34 billion proteins… then you’d just use a hidden Markov model and match… Cas9”
MB
Mammoth Biosciences
The Neuron: AI Explained ·
“We will license core ai models from really deep resourced company and then train those models on client data ... without having to hire 300 people”
♫
Said on InsTech - insurance & innovation
Moments like this, every Monday
The strongest new moments on your topics, in the speakers' own words. Free for three topics.
No card. No password. One-click unsubscribe.
Track a competitor?
Pro watches up to 25 names or phrases and emails you the morning after they are said.
See Pro, $29 a monthVerbatim quotes a short extract and links to the episode; the episode belongs to its publisher. Remove a show.